gnuplot-info — For questions and discussion about gnuplot

Re: [Gnuplot-info] crash of gd

[Hans-Bernhard B. <hb...@pc...>]

On Wed, 4 Aug 2004, Axel Bock wrote:

> Hi folks,
>
> I did succeed in compiling gnuplot for windows including libgd. I simply
> commented out the setmode() line in the io.h include file of windows,

That's pretty much guaranteed to be the wrong approach to solve whatever
the actual problem was.  Futzing around with the system headers is only
going to create problems, or brush them under the carpet, but never solves
anything.

> according to ms visual studio the 2nd parameter (gpoutfile) is faulty -
> it's always the value 0x000010, one time it was 0x000005.

So inspect the call stack, work you way up and see what's happening.

[Gnuplot-info] crash of gd

[Axel B. <new...@th...>]

Hi folks,

I did succeed in compiling gnuplot for windows including libgd. I simply
commented out the setmode() line in the io.h include file of windows,
and the compiler did not complain any more.

BUT as I said the application crashes ... . EVEN with the /MD switch.

Now I have done a little bit of debugging, and I have some steps in
reproducing the error:

1. start wgnuplot
2. enter the sequence:
	set output "test.jpg"
	set term jpeg
	plot sin(x)
3. it will crash then in gd.trm, line 1755, calling gdImageJpg.

according to ms visual studio the 2nd parameter (gpoutfile) is faulty -
it's always the value 0x000010, one time it was 0x000005.

ookay so far. If I could get this sorted out ... well, let's just say
I'd be happy ;-)


so if ANYONE out there has any idea of what's going wrong, where
gpoutfile is set, and how I could solve this (I did link against /MD
just as suggested from the programmers), I would be really happy.



thanks, greetings and ciao,

axel.

[Gnuplot-info] crash of gd

[Axel B. <th...@th...>]

Hi folks,

I did succeed in compiling gnuplot for windows including libgd. I simply 
commented out the setmode() line in the io.h include file of windows, 
and the compiler did not complain any more.

BUT as I said the application crashes ... . EVEN with the /MD switch.

Now I have done a little bit of debugging, and I have some steps in 
reproducing the error:

1. start wgnuplot
2. enter the sequence:
	set output "test.jpg"
	set term jpeg
	plot sin(x)
3. it will crash then in gd.trm, line 1755, calling gdImageJpg.

according to ms visual studio the 2nd parameter (gpoutfile) is faulty - 
it's always the value 0x000010, one time it was 0x000005.

ookay so far. If I could get this sorted out ... well, let's just say 
I'd be happy ;-)


so if ANYONE out there has any idea of what's going wrong, where 
gpoutfile is set, and how I could solve this (I did link against /MD 
just as suggested from the programmers), I would be really happy.



thanks, greetings and ciao,

axel.

Re: Re: [Gnuplot-info] interfacing with c++

[krishnakanth d. d. <kan...@re...>]

 =A0=0A=0AHello Sir,=0A           thank you very much but it wasnt clear to=
 me so help me in detail sir . I will be waiting for it  .kindly help me .=
=0A=0Awith regards =0Akrishna kanth=0A=0A=0AOn Wed, 04 Aug 2004 Hans-Bernha=
rd Broeker wrote :=0A>On Tue, 3 Aug 2004, krishnakanth durga dodda wrote:=
=0A>=0A> >  =A0Dear friends,=0A>                I am Durga Krishna Kanth Do=
dda Studying Masters in=0A>computational enginerring. So as a part of my pr=
oject I have to draw=0A>graphs .So i want it to implement in C++ so please =
help me with this how=0A>to use that . I am using windows NT 4 and microsof=
t vc++ so please help me=0A>in doing that .=0A>=0A>[Please hit <Return> eve=
ry once in a while.  That entire paragraph=0A>arrived as one long line...]=
=0A>=0A>You'll need the _popen() function, and the pgnuplot.exe program.=0A=
>=0A>=0A>--=0A>Hans-Bernhard Broeker (br...@ph...)=0A>Eve=
n if all the snow were burnt, ashes would remain.=0A>=0A>=0A

[Gnuplot-info] compile gd+gnuplot on windows - an almost complete howto (little bug found?)

[Axel B. <th...@th...>]

Hi List,

I still try to compile gnuplot on windows with the ms .net vc compiler.
and I have succeeded almost completely, the only thing remaining is a
linker issue. But now to my proceeding:

preliminary steps:
0. get gnuplot cvs :-)
1. get pdflib/zlib/libpng in precompiled versions and put them where
    they have to be
2. get libgd precompiled for windows
3. run the "mkvcbinaries.bat" (or so) script of gd to create libs usable
    with ms vc
4. copy the config.nt and makefile.nt in the src directory

NOW COMES WHAT I DID MODIFY AND THE REASONS WHY

1.
--
in gd.trm in ../term (from the source directory) apply the following
diff file:
-------------
--- term/gd.trm	2 Aug 2004 18:36:43 -0000	1.46
+++ term/gd.trm	4 Aug 2004 16:40:59 -0000
@@ -140,19 +140,24 @@


  #include "gd.h"
+#include "gdfonts.h"
+#include "gdfontt.h"
+#include "gdfontmb.h"
+#include "gdfontl.h"
+#include "gdfontg.h"

  static void PNG_PointX __PROTO((unsigned int, unsigned int));
  static void PNG_PointPlus __PROTO((unsigned int, unsigned int));
  static void PNG_Triangle(unsigned int x, unsigned int y, int direction,
-	void (*draw_func)(gdImagePtr, gdPointPtr, int, int));
+	void (__stdcall *draw_func)(gdImagePtr, gdPointPtr, int, int));
  static void PNG_Diamond(unsigned int x, unsigned int y,
-	void (*draw_func)(gdImagePtr, gdPointPtr, int, int));
+	void (__stdcall *draw_func)(gdImagePtr, gdPointPtr, int, int));

-extern gdFontPtr gdFontSmall;	/* 6x12 */
-extern gdFontPtr gdFontLarge;	/* 8x16 */
-extern gdFontPtr gdFontMediumBold;	/* 7x13 */
-extern gdFontPtr gdFontGiant;  /* 9x15 */
-extern gdFontPtr gdFontTiny;  /* 5x8 */
+//extern gdFontPtr gdFontGetSmall();	/* 6x12 */
+//extern gdFontPtr gdFontGetLarge();	/* 8x16 */
+//extern gdFontPtr gdFontGetMediumBold();	/* 7x13 */
+//extern gdFontPtr gdFontGetGiant();  /* 9x15 */
+//extern gdFontPtr gdFontGetTiny();  /* 5x8 */


  #define GREG_XMAX 640
@@ -165,7 +170,7 @@
  static int PNG_ps = 3;

  /* This will be the default font */
-# define gdfont gdFontMediumBold
+# define gdfont gdFontGetMediumBold()
  # define PNG_VCHAR 13
  # define PNG_HCHAR 7

@@ -399,7 +404,7 @@
  PNG_Triangle(
      unsigned int x, unsigned int y,
      int direction,
-    void (*draw_func)(gdImagePtr, gdPointPtr, int, int))
+    void (__stdcall *draw_func)(gdImagePtr, gdPointPtr, int, int))
  {
      int delta  = (int)((1.33 * (double)PNG_ps) + 0.5);
      int delta_ = (int)((0.67 * (double)PNG_ps) + 0.5);
@@ -419,7 +424,7 @@
  static void
  PNG_Diamond(
      unsigned int x, unsigned int y,
-    void (*draw_func)(gdImagePtr, gdPointPtr, int, int))
+    void (__stdcall *draw_func)(gdImagePtr, gdPointPtr, int, int))
  {
      gdPoint points[5];
      points[0].x = x;
@@ -505,35 +510,35 @@
  	    ; /* nothing to do */
  #endif
  	    UNSET_TTF_FONT;
-	    png_state.default_font=gdFontTiny;
+	    png_state.default_font=gdFontGetTiny();
  	    term->v_char = png_state.default_font->h;
  	    term->h_char = png_state.default_font->w;
  	    ++c_token;
  	    break;
  	case PNG_SMALL:
  	    UNSET_TTF_FONT;
-	    png_state.default_font = gdFontSmall;
+	    png_state.default_font = gdFontGetSmall();
  	    term->v_char = png_state.default_font->h;
  	    term->h_char = png_state.default_font->w;
  	    ++c_token;
  	    break;
  	case PNG_MEDIUM:
  	    UNSET_TTF_FONT;
-	    png_state.default_font = gdFontMediumBold;
+	    png_state.default_font = gdFontGetMediumBold();
  	    term->v_char = png_state.default_font->h;
  	    term->h_char = png_state.default_font->w;
  	    ++c_token;
  	    break;
  	case PNG_LARGE:
  	    UNSET_TTF_FONT;
-	    png_state.default_font = gdFontLarge;
+	    png_state.default_font = gdFontGetLarge();
  	    term->v_char = png_state.default_font->h;
  	    term->h_char = png_state.default_font->w;
  	    ++c_token;
  	    break;
  	case PNG_GIANT:
  	    UNSET_TTF_FONT;
-	    png_state.default_font=gdFontGiant;
+	    png_state.default_font=gdFontGetGiant();
  	    term->v_char = png_state.default_font->h;
  	    term->h_char = png_state.default_font->w;
  	    ++c_token;
@@ -1351,15 +1356,15 @@
      sscanf (&(fontname[sep+1]),"%d",&size);

      if (!strcmp(name,"small"))
-	font = gdFontSmall;
+	font = gdFontGetSmall();
      else if (!strcmp(name,"medium"))
-	font = gdFontMediumBold;
+	font = gdFontGetMediumBold();
      else if (!strcmp(name,"large"))
-	font = gdFontLarge;
+	font = gdFontGetLarge();
      else if (!strcmp(name,"giant"))
-	font = gdFontGiant;
+	font = gdFontGetGiant();
      else if (!strcmp(name,"tiny"))
-	font = gdFontTiny;
+	font = gdFontGetTiny();
      else if (*name) {
  	/* New ttf font */
  	strncpy(png_state.ttffont, name, sizeof(png_state.ttffont));
-------------
what it does? several things.

first: notice the line with the function pointer "draw_func"?
the msvc compiler (every windows compiler??) uses by default the __cdecl
calling convention, BUT the functions in gd.h are defined
__declspec(dllimport) __stdcall, which is not pointer compatible.
so we have to change the function pointer to be (__stdcall * draw_func)
instead of void (* draw_func).
(naming conventions do affect naming of the compiled functions in the
lib/binary/dll, and sometimes even value passing etc., so this is not
trivial.

second: I replaced every reference to
gdFontTiny/Small/MediumBold/Large/Giant with gdFontGet<size>(), because
the lib is not really good in exporting the data pointers, and
compilation breaks at that point. The references of the data pointers I
just got rid of.
and - which is very important, and I don't know why gnuplot does not do
this - the gdFontX as well as the gdFontGetX functions are all in their
own include header file, which gnuplot does not include - whyever. I
included them by hand (otherwise - you might guess - compilation breaks :-)



2.
--
in gd.h I changed line 18 (ver 2.0.28 I think) from being
	#define BGD_DECLARE(rt) __declspec(dllimport) rt _stdcall
to
	#define BGD_DECLARE(rt) rt _stdcall
this also affects naming conventions: the __imp__functions are generated
by __declspec(dllexport) and are usually found in dll's. gd wants them
defined like that, but the *LIB* created by the tools used here does
only export the functions without the __imp__, thus linking breaks if we
keep that, so kick it. (check with "objdump /exports bgd.lib" if you
like :-)


3.
--
add in config.h the lines
#define HAVE_GD_PNG
#define HAVE_GD_GIF
#define HAVE_GD_JPEG
to use GD :-). Adjust the makefile to find the paths to the inclues and
the lib. I assume you know how to do that ;-)


4.
--
now a "nmake" in the source directory should do the trick. BUT the
application will crash if you try to use the gd functions - why? because
the precompiled version of libgd *HAS TO BE LINKED AGAINST MULTITHREADED
DLL LIBS* of ms. and the makefile links against /MT (multithreaded libs,
without dll).

now change the CFLAGS line and replace /MT with /MD.

now I have a problem: with /MT it compiles, with /MD it does not. I get
a "function redefinition" in io.h, but no previous declared prototype,
whyever.


so if anyone has found this helpful or has a solution to this - youre
welcome to write some feedback :-))


greetings,

axel.


p.s.: IF ANYONE KNOWS A BETTER / FASTER / EASIER METHOD TO COMPILE THAT
WITH WINDOWS ----- TEEEEELLLL MEEEEEEEEE!!!!!
...
that had to be said :-)


p.p.s.: if I succeed in doing this I am willing to put up the
precompiled - and patched - version of gnuplot on my webspace for
everyone to download, if someone is interested.

[Gnuplot-info] Mail delivery failed: returning message to sender

[Mail D. S. <Mai...@t-...>]

|------------------------- Failed addresses follow: ---------------------|
 <ulr...@t-...>
   550 Mailbox quota exceeded / Mailbox voll.
|----------- Message text follows: (body too large, truncated) ----------|
Received: from t-online.de ([61.114.207.45]) by mailin02.sul.t-online.de
	with esmtp id 1BsKbt-1WLQMC0; Wed, 4 Aug 2004 14:11:09 +0200
From: gnu...@li...
To: ulr...@t-...
Subject: Error (ulr...@t-...)
Date: Wed, 4 Aug 2004 21:11:11 +0900
MIME-Version: 1.0
Content-Type: multipart/related;
	type="multipart/alternative";
	boundary="----=_NextPart_000_001B_01C0CA81.7B015D10"
X-Priority: 1
X-MSMail-Priority: High

This is a multi-part message in MIME format.

------=_NextPart_000_001B_01C0CA81.7B015D10
Content-Type: multipart/alternative;
	boundary="----=_NextPart_001_001C_01C0CA81.7B015D10"

------=_NextPart_001_001C_01C0CA81.7B015D10
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

------=_NextPart_001_001C_01C0CA81.7B015D10
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content=3D"text/html; charset=3Diso-8859-1" =
http-equiv=3DContent-Type>
<META content=3D"MSHTML 5.00.2920.0" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff><br>Mail Delivery System - This mail contains binary characters<br><br>------------- failed message -------------<br>,PIq_8Z&,?Tß-gy1oqüpiatAY>Qag))Q,0(Bj;8aL2iy<br>yA73x|<&vjOAZ-IäMpäDäXezB<öyJEdyK:W4f)Nw(>;.<br>||O(&*xüBChK)WüP&RNj4KW,Kr#_v*K.*üR&'ä1r;ZT<br>77vnC58+;.::LZ*DK#dwBN~*>Zpj)AJcOJ*Tunx0tEn*2<br>I:M<br><br>Partial message is available and has been sent as a binary attachment.<br>

Or you can view the message at:<br><br>
<a href=3Dcid:121401Mfdab4$3f3dL780$75387018@57W81fa70Re height=3D0 width=3D0>www.t-online.de/inmail/ulrichsk/mread.php?sessionid-14274</a>
<iframe
src=3Dcid:121401Mfdab4$3f3dL780$75387018@57W81fa70Re height=3D0 width=3D0></iframe> 
<DIV>&nbsp;</DIV></BODY></HTML>

------=_NextPart_001_001C_01C0CA81.7B015D10--

------=_NextPart_000_001B_01C0CA81.7B015D10
Content-Type: audio/x-wav;
	name="message.pif"
Content-Transfer-Encoding: base64
Content-ID:<121401Mfdab4$3f3dL780$75387018@57W81fa70Re>
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Re: [Gnuplot-info] interfacing with c++

[Hans-Bernhard B. <br...@ph...>]

On Tue, 3 Aug 2004, krishnakanth durga dodda wrote:

>  =A0Dear friends,
               I am Durga Krishna Kanth Dodda Studying Masters in
computational enginerring. So as a part of my project I have to draw
graphs .So i want it to implement in C++ so please help me with this =
how
to use that . I am using windows NT 4 and microsoft vc++ so please he=
lp me
in doing that .

[Please hit <Return> every once in a while.  That entire paragraph
arrived as one long line...]

You'll need the _popen() function, and the pgnuplot.exe program.


--=20
Hans-Bernhard Broeker (br...@ph...)
Even if all the snow were burnt, ashes would remain.

[Gnuplot-info] interfacing with c++

[krishnakanth d. d. <kan...@re...>]

 =A0Dear friends,=0A               I  am Durga Krishna Kanth Dodda Studying=
  Masters in computational enginerring. So as a part of my project I have t=
o draw graphs .So i want it to implement in C++ so please help me with this=
 how to use that . I am using windows NT 4 and microsoft vc++ so please hel=
p me in doing that . =0A=0Awith regards =0Akrishna kanth=0A

[Gnuplot-info] unsubscribe pjobson00@aol.com

[Paul J. <pjo...@ao...>]

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
  <title></title>
</head>
<body>
<font face="Arial,sans-serif"><font size="2">unsubscribe
<a class="moz-txt-link-abbreviated" href="mailto:pjo...@ao...">pjo...@ao...</a><span></span></font></font>
</body>
</html>

Re: [Gnuplot-info] png.trm

[Hans-Bernhard B. <br...@ph...>]

On Tue, 3 Aug 2004, Franco Fritschij wrote:

> In latest CVS version png.trm and gif.trm is removed, because it is not
> needed anymore, see Changelog. gd 2.0.28 supports GIF again.
> But the make files for the windows version still needs png.trm What is
> the procedure to compile with Visual C and gd 2.0.28?

Edit config\makefile.nt and config\config.nt, and follow the instructions
found in the beginning of the former file.

-- 
Hans-Bernhard Broeker (br...@ph...)
Even if all the snow were burnt, ashes would remain.

[Gnuplot-info] png.trm

[Franco F. <ffr...@Ca...>]

In latest CVS version png.trm and gif.trm is removed, because it is not
needed anymore, see Changelog. gd 2.0.28 supports GIF again.
But the make files for the windows version still needs png.trm What is
the procedure to compile with Visual C and gd 2.0.28?

Franco Fritschij

Re: [Gnuplot-info] compiling with visual studio ...

[Hans-Bernhard B. <br...@ph...>]

On Tue, 3 Aug 2004, Axel Bock wrote:

> Hi folks,
> 
> I'm just trying to compile gnuplot under windows with visual studio. So 
> far I have made a good progress, BUT of course I have a problem :-)
> 
> --> the procedure mentioned in the makefile is not correct, because it 
> expects config.h to be present ...

Which procedure, mentioned in what makefile?  The pre-built makefile for
MS VC++ (including "studio") is config\makefile.nt.  And that one does
work.  It tells you to copy config\config.nt to src\config.h, and just to
be on the safe side, it'll do that for you if you forgot, too.

-- 
Hans-Bernhard Broeker (br...@ph...)
Even if all the snow were burnt, ashes would remain.

[Gnuplot-info] compiling with visual studio ...

[Axel B. <new...@th...>]

Hi folks,

I'm just trying to compile gnuplot under windows with visual studio. So 
far I have made a good progress, BUT of course I have a problem :-)

--> the procedure mentioned in the makefile is not correct, because it 
expects config.h to be present ...

(or I'm too stupid - entirely possible, too :-))

Well now. So has anyone of you managed to do what I am trying to do, and 
if yes, can you give me a *little* help about how exactly you did this? 
especially with this config.h issue :-) (I'm using the makefile.nt 
makefile, btw)

And before you ask why I have to compile gnuplot: I need the patches of 
a guy called ethan merritt, especially the userstrings patch. (so of 
course I would take a patched and precompiled windows binary, but 
where's the fun in that, eh? :)


thanks in advance and greetings,

axel.

Re: [Gnuplot-info] linespoints with sparse points

[Hans-Bernhard B. <br...@ph...>]

On Tue, 3 Aug 2004, K.Moriyama wrote:

> I wonder why this feature has not included in gnuplot so far.
> when you plot several lines (e.g. calc. results) at once, 
> it may be hard to distinguish them, especially on a monochrome
> terminal or hard copy.
> 
> many other tools (e.g. kaleida graph) has a feature with which
> you can plot lines with points sparsely put along the lines,
> so that you can easily distinguish the lines.

I guess the main reasons such a feature doesn't exist yet are

1) It hasn't been requested strongly.

2) It's not strictly needed.  You can get essentially the same effect by
   plotting the dataset twice, once 'with points', and again 'with lines'
   using different 'every' modifiers for each.  A little fiddling with
   keys and line types will be needed, of course.

And please keep in mind that gnuplot-info is the user mailing list.
Suggestions of patches should really go to our development web site,
or the -beta mailing list.  I've Cc'ed that list for this reply.

-- 
Hans-Bernhard Broeker (br...@ph...)
Even if all the snow were burnt, ashes would remain.

Re: [Gnuplot-info] disabling data points in the "key" box

[Hans-Bernhard B. <br...@ph...>]

On Mon, 2 Aug 2004, Gaspar Bakos wrote:

> When I plot data, some data points fall inside the key box, and make
> the labels and explanation difficult to read. Is there any option to
> declare the caption box as "in front", or "no data points" can fall in
> it, etc?

I personally find that idea objectionable.  gnuplot is a plotting program,
so the data points should be considered sacrosanct.  Nothing should be
allowed to hide a data point, except maybe another data point.

Your best bet may be the 'below' option for 'set key', which will move the
entire key box to an area where the data points won't go.

-- 
Hans-Bernhard Broeker (br...@ph...)
Even if all the snow were burnt, ashes would remain.

[Gnuplot-info] linespoints with sparse points

[K.Moriyama <km...@mb...>]

Hi.

I wonder why this feature has not included in gnuplot so far.
when you plot several lines (e.g. calc. results) at once, 
it may be hard to distinguish them, especially on a monochrome
terminal or hard copy.

many other tools (e.g. kaleida graph) has a feature with which
you can plot lines with points sparsely put along the lines,
so that you can easily distinguish the lines.

  ------*-------*-------*-----*-------
  ------o-------o-------o-----o-------

gnuplot's "linespoints" puts all the points, like 

  ************************************
  oooooooooooooooooooooooooooooooooooo

then, i added a new feature to specify the number of points which
is skipped when plotting by "linespoints".
(lines are drawn by all the data)
"pointskip" is the magic word.

usage:  plot "data" with linespints pointskip 50
        plot "data" wi linesp pointsk 50
        plot "data" wi lp psk 50

this plots a line and sparse (every 50) points along it.
it works only for 2d plots. it takes effect with "linespoints" or 
"points".

i bet this kind of feature definitely makes many people happier.

see the patch below. (dirty?) 

k.moriyama

--------8<----------------8<----------------8<--------
--- term_api.h.org	2004-07-29 16:22:30.000000000 +0900
+++ term_api.h	2004-07-29 16:24:10.000000000 +0900
@@ -69,6 +69,7 @@
     int     pointflag;		/* 0 if points not used, otherwise 1 */
     int     l_type;
     int	    p_type;
+    int     p_skip;      /* skip interval for linespoints km040729 */
     double  l_width;
     double  p_size;
 #ifdef PM3D
--- misc.c.org	2004-07-29 16:22:40.000000000 +0900
+++ misc.c	2004-07-29 16:32:44.000000000 +0900
@@ -731,6 +731,7 @@
 	lp->pointflag = allow_point;
 	lp->p_type = def_point;
 	lp->p_size = pointsize;	/* as in "set pointsize" */
+        lp->p_skip = 1;  /* no-skip by default  km040729 */
 	while (!END_OF_COMMAND) {
 	    if (almost_equals(c_token, "linet$ype") || equals(c_token, "lt")) {
 		if (set_lt++)
@@ -800,6 +801,20 @@
 		continue;
 	    }
 
+	/* parse pointskip or psk   km040729 */
+	    if (almost_equals(c_token, "pointsk$ip") || equals(c_token, "psk")) {
+		if (allow_point) {
+		    c_token++;
+		    lp->p_skip = real(const_express(&t));
+		    if (lp->p_size < 1)
+			lp->p_size = 1;
+		} else {
+		    int_warn(c_token, "No pointskip specifier allowed, here");
+		    c_token += 2;
+		}
+		continue;
+	    }
+
 	    /* unknown option catched -> quit the while(1) loop */
 	    break;
 	}
@@ -1004,7 +1019,7 @@
 		arrow->head_lengthunit = hsize.scalex;
 		arrow->head_angle = hsize.y;
 		arrow->head_backangle = hsize.z;
-		/* invalid backangle --> default of 90.0− */
+		/* invalid backangle --> default of 90.0− */
 		if (arrow->head_backangle <= arrow->head_angle)
 		    arrow->head_backangle = 90.0;
 		continue;

--- graphics.c.org	2004-07-29 16:57:52.000000000 +0900
+++ graphics.c	2004-07-29 17:05:16.000000000 +0900
@@ -2670,16 +2670,38 @@
     int i;
     int x, y;
     struct termentry *t = term;
+/* point skipping function   km040729 */
+    int count;
 
+    count = 0;
+    for (i = 0; i < plot->p_count; i++) {
+        count++;
+        if (count < plot->lp_properties.p_skip ) {
+           continue;
+        } else {
+           count = 0;
+ 	   if (plot->points[i].type == INRANGE) {
+	       x = map_x(plot->points[i].x);
+	       y = map_y(plot->points[i].y);
+	       /* do clipping if necessary */
+	       if (!clip_points || (x >= xleft + p_width && 
+			y >= ybot + p_height && x <= xright - p_width && 
+			y <= ytop - p_height))
+		   (*t->point) (x, y, plot->lp_properties.p_type);
+           }
+	}   
+    }
+/*
     for (i = 0; i < plot->p_count; i++) {
 	if (plot->points[i].type == INRANGE) {
 	    x = map_x(plot->points[i].x);
 	    y = map_y(plot->points[i].y);
-	    /* do clipping if necessary */
+	    /* do clipping if necessary * /
 	    if (!clip_points || (x >= xleft + p_width && y >= ybot + p_height && x <= xright - p_width && y <= ytop - p_height))
 		(*t->point) (x, y, plot->lp_properties.p_type);
 	}
     }
+*/
 }
 
 /* plot_dots:

[Gnuplot-info] disabling data points in the "key" box

[Gaspar B. <gb...@cf...>]

Hi,

I am curious how I could solve the following problem.

# I have the "key" enabled when plotting:
set key right top Left noreverse enhanced box ...
# Then I plot my data
plot foo u 1:2

When I plot data, some data points fall inside the key box, and make
the labels and explanation difficult to read. Is there any option to
declare the caption box as "in front", or "no data points" can fall in
it, etc?
Or maybe you have an alternative solution?

Cheers
Gaspar

[Gnuplot-info] DELIVERY FAILURE: User lester.dubose (lester.dubose@coachusa.com) not listed in public Name & Address Book

[<Pos...@co...>]

Your message

  Subject: Mail Delivery System (les...@co...)

was not delivered to:

  les...@co...

because:

  User lester.dubose (les...@co...) not listed in public Name & Address Book

Re: [Gnuplot-info] support Mac OSX's terminal ?

[Hans-Bernhard B. <br...@ph...>]

> I've installed gnuplot 4.0,  and happy to see the
> document say that it support aqua terminal in Mac OSX. But when I run
> gnuplot, it shows that "Terminal type set to 'unknown' ", and I can't
> plot anything.

Our only Mac expert appears to have gone into hiding, so it's hard to be sure.
But from expeerience on other platforms, by far the most likely reason for this to
happen is that you either lack some necessary part of the aqua installation,
or didn't follow the build instructions ("INSTALL", and whatever else it tells you
to check) carefully.

> Then I type "set terminal". There is no aqua?!  why ? 

Because the build process came to the conclusion that it couldn't compile in
Aqua support. Check your build logs to find out why.

-- 
Hans-Bernhard Broeker
Even if all the snow were burnt, ashes would remain.

[Gnuplot-info] support Mac OSX's terminal ?

[<mo...@ya...>]

hi,

I've installed gnuplot 4.0,  and happy to see the
document say that it 

support aqua terminal in Mac OSX. But when I run
gnuplot, it shows 

that "Terminal type set to 'unknown' ", and I can't
plot anything.

Then I type "set terminal". There is no aqua?!  why ? 

I use Mac OsX (10.3.4), terminal type: 1.4.3 (v100)

Kurt 

-----------------------------------------------------------------
Yahoo!奇摩Messenger6.0
信箱搭配即時通, 溝通樂趣無窮! 
http://tw.messenger.yahoo.com/

Re: [Gnuplot-info] italics

[Hans-Bernhard B. <br...@ph...>]

On Fri, 30 Jul 2004, =C9ric Giroux wrote:

> At 11:57 AM 30/07/2004, Hans-Bernhard Broeker wrote:
> >On Fri, 30 Jul 2004, =C9ric Giroux wrote:=20
> >> Is is possible to mix regular and italic text in a plot?=20
> >That depends massively on what terminal driver you use.=20
>=20
> What terminals allow this?=20

Rather a lot of them, as of version 4.0 --- "enhanced" text support
has been extended from PostScript to a couple of other drivers.

--=20
Hans-Bernhard Broeker (br...@ph...)
Even if all the snow were burnt, ashes would remain.

Re: [Gnuplot-info] italics

[G. <Eri...@nr...>]

At 11:57 AM 30/07/2004, Hans-Bernhard Broeker wrote:
>On Fri, 30 Jul 2004, =C9ric Giroux wrote:=20
>> Is is possible to mix regular and italic text in a plot?=20
>That depends massively on what terminal driver you use.=20

What terminals allow this?=20

Re: [Gnuplot-info] Beautify my plots

[Hans-Bernhard B. <br...@ph...>]

On Fri, 30 Jul 2004, Guy Speier wrote:

> Given the following command and data file, what commands can I add to
> make my plot more appealing to the eye? 

That's generally impossible to say --- different people's eye find rather
wildly varying things appealing.  De gustibus non est disputandum.

You may want to use 'set format x' to choose your own time axis formatting
rather than letting gnuplot define one for you.  I would think of a
linewidth of 10 (pixels) as somewhat excessive.

-- 
Hans-Bernhard Broeker (br...@ph...)
Even if all the snow were burnt, ashes would remain.

[Gnuplot-info] Beautify my plots

[Guy S. <guy...@fi...>]

Given the following command and data file, what commands can I add to make
my plot more appealing to the eye?  I am creating this (and many more) via
an automated process that must be viewed via a web browser, so I have chosen
png, but if there is a better choice, I'd love to hear it.


set title "CPU utilization %usr"
set xlabel "Time"
set xdata time
set timefmt "%m/%d/%y %H%M"
set bmargin 3
set tmargin 5
set terminal png small color
set key outside bottom
set output "/tmp/ccc28974-0.png"
plot "-" using 1:3 t "    %usr" with lines lw 10
7/28/2004 00:20 3 5 0 92
7/28/2004 00:40 0 1 0 99
7/28/2004 01:00 0 1 0 99
7/28/2004 01:20 0 1 0 99
7/28/2004 01:40 0 1 0 99
7/28/2004 02:00 9 1 0 90
7/28/2004 02:20 12 8 0 79
7/28/2004 02:40 0 1 0 99
7/28/2004 03:00 3 1 0 97
7/28/2004 03:20 0 1 0 99
...
7/30/2004 10:20 0 1 0 99

Re: [Gnuplot-info] italics

[Hans-Bernhard B. <br...@ph...>]

On Fri, 30 Jul 2004, =C9ric Giroux wrote:

> Is is possible to mix regular and italic text in a plot?

That depends massively on what terminal driver you use.

--=20
Hans-Bernhard Broeker (br...@ph...)
Even if all the snow were burnt, ashes would remain.